---
type: "Article"
title: "Qwen3.8 27B EXL3 4 bpw with ExLlamaV3 on 1× RTX 5090 (one-click kit)"
description: "Qwen3.8 27B EXL3 4 bpw with ExLlamaV3 on 1× RTX 5090 (one-click kit): recipe by MiaAI-Lab with speed, condition and source for every number, weights and license."
resource: "https://www.contextstudios.ai/local-ai/miaai-lab--qwen3.8-27b-exl3-rtx-5090-one-click"
language: "en"
generated:
  by: "process:contextstudios-md/1"
  at: "2026-10-24T16:07:29.439Z"
status: "stable"
---

# Qwen3.8 27B EXL3 4 bpw with ExLlamaV3 on 1× RTX 5090 (one-click kit)

Qwen3.8 27B EXL3 4 bpw with ExLlamaV3 on 1× RTX 5090 (one-click kit): 122 tok/s according to github.com (dataset as of Sep 29, 2026).

- Creator: [MiaAI-Lab](https://www.contextstudios.ai/local-ai/creators/miaai-lab.md)
- Engine: ExLlamaV3 v1.6.0 (tools/serve_openai.py), MTP; opt-in DRAFT=dflash2
- Quantization: EXL3 4.0 bpw (turboderp) + int4 KV
- Model family: Qwen3.8
- Hardware: rtx-5090 x1
- Context: 262144
- Artificial Analysis Intelligence Index: 33.7 ([source](https://artificialanalysis.ai/models/qwen3-8-27b))

## All numbers

- 122 tok/s (Mixed; Mixed, 1 request, llama-benchy 2048-token prompt, 128 tokens out, depth 0, with MTP) — [source](https://github.com/MiaAI-Lab/Qwen3.8-27B-16gb-NVIDIA-GPUs-one-click-install#what-was-measured)
- decode_tps: 121.8 (1 request) — [source](https://github.com/MiaAI-Lab/Qwen3.8-27B-16gb-NVIDIA-GPUs-one-click-install#what-was-measured)
- decode_tps: 366.4 (1 request) — [source](https://github.com/MiaAI-Lab/Qwen3.8-27B-16gb-NVIDIA-GPUs-one-click-install#what-was-measured)

## Weights

- [turboderp/Qwen3.8-27B-exl3](https://huggingface.co/turboderp/Qwen3.8-27B-exl3) (main, apache-2.0)
- [r0b0tlab/Qwen3.8-27B-DFlash2-EXL3-4.00bpw](https://huggingface.co/r0b0tlab/Qwen3.8-27B-DFlash2-EXL3-4.00bpw) (draft, apache-2.0)
- [Qwen/Qwen3.8-27B](https://huggingface.co/Qwen/Qwen3.8-27B) (base, apache-2.0)
- [Mia-AiLab/Qwen3.8-27B-EXL3-2.0bpw](https://huggingface.co/Mia-AiLab/Qwen3.8-27B-EXL3-2.0bpw) (other, apache-2.0)

[https://github.com/MiaAI-Lab/Qwen3.8-27B-16gb-NVIDIA-GPUs-one-click-install](https://github.com/MiaAI-Lab/Qwen3.8-27B-16gb-NVIDIA-GPUs-one-click-install)

## Related

- [Local AI on your hardware.](https://www.contextstudios.ai/local-ai.md)